EC, July 14, 2010, No M.5747

Dear Sir/Madam,

Subject: Case No COMP/M.5747 – Iberia/ British Airways
Notification of 10 June 2010 pursuant to Article 4 of Council Regulation No 139/20041

1. On 10 June 2010, the Commission received a notification of a proposed concentration pursuant to Article 4 of the Merger Regulation by which the undertakings Iberia Líneas Aéreas de España, S.A. ("Iberia" or "IB", Spain) and British Airways Plc ("British Airways" or "BA", United Kingdom) enter into a full merger within the meaning of Article 3(1)(a) of the Merger Regulation.

I.THE PARTIES

2. British Airways serves some 150 cities in about 75 countries with an additional 215   cities served under various code-sharing relationships. It has several subsidiaries (CityFlyer, OpenSkies) and franchisees (Comair and Sunair). BA has its main hub at London Heathrow airport and also operates bases at London Gatwick and London City. In addition to passenger transport services, BA provides air transport of cargo services, aircraft maintenance and groundhandling services.

3. Iberia serves more than 106 destinations in over 43 countries with an additional 110 cities served under various code-sharing relationships. Its main hubs are in Madrid and Barcelona. IB's operations in Barcelona are carried out mainly through its subsidiary Vueling. IB's franchisee Air Nostrum operates mostly on Spanish domestic routes. In addition to passenger transport services, IB provides air transport of cargo services, aircraft maintenance and groundhandling services.

4. Both BA and IB are members of the oneworld alliance.2

II. CONCENTRATION 

5. On 8 April 2010, IB and BA entered into a merger agreement according to which they will integrate the entirety of their activities. Pursuant to the merger agreement, a recently incorporated Spanish company TopCo, will become the holding company of both airlines that will remain separate entities operating under their respective trade names. TopCo's shareholders will be the existing shareholders of IB and BA.

6. The proposed transaction thus constitutes a merger within the meaning of Article 3(1)(a) of the Merger Regulation.

III. EU DIMENSION

7. The undertakings concerned, BA and IB, have a combined aggregate world-wide turnover of more than EUR 5 000 million.3 Each of them has an EU-wide turnover in excess of EUR 250 million, but they do not achieve more than two-thirds of their aggregate EU-wide turnover within one and the same Member State. The methodology used by the notifying parties to calculate their turnover is the "point of sale" methodology, although in any event the thresholds would also be met under the "point of origin" or "50/50 split" methods.4 The notified concentration therefore has an EU dimension.

IV. RELEVANT PRODUCT AND GEOGRAPHIC MARKETS

8. The activities of IB and BA overlap in air transport of passengers, air transport of cargo, groundhandling services and maintenance, repair, and overhaul services ("MRO").

A. Passenger air transport services

A.1  O&D approach

9. The Commission has traditionally defined the relevant market for scheduled passenger air transport services on the basis of the "point of origin/point of destination" (O&D) city-pair approach.5 Such market definition reflects the demand-side perspective whereby customers consider all possible alternatives of travelling from a city of origin  to a city of destination which they do not consider substitutable to a different city-pair. As a result, every combination of a point of origin and a point of destination is considered to be a separate market.

10. The market investigation has confirmed this approach in the present case. While in particular the parties and some network carriers tend to believe that the market should  be defined on a network basis, the O&D approach best reflects the demand-side perspective according to which both the point of origin and the point of destination should include all airports that are substitutable in the eyes of passengers.

A.2 Time sensitive vs. non-time sensitive passengers

11. The Commission has traditionally found that a distinction may be drawn between time sensitive and non-time sensitive passengers. Time sensitive passengers tend to travel for business purposes, require significant flexibility with their tickets (such as cost-free cancellation and modification of the time of departure, etc.) and tend to pay higher  prices for this flexibility. Non-time sensitive customers travel predominantly for leisure purposes or to visit friends and relatives, book long time in advance, do not require flexibility with their booking and are generally more price-sensitive.6

12.The parties submit that such a distinction is not meaningful as a chain of substitution (between fare classes or cabins) exists between the two categories of passengers.

13. The market investigation however showed that time sensitive passengers often have different requirements from non-time sensitive passengers, in particular as regards flexibility, the level of frequencies, convenient schedules, airport location, and comfort. Accordingly, time sensitive passengers are willing to pay a significant premium for such features.7

14. A relevant number of respondents amongst the parties' competitors had moreover indicated that airlines are able to identify and discriminate passengers according to their price elasticity based on their observable behaviour at the time of booking. In particular, passengers with lower price elasticity tend to purchase tickets with a high degree of flexibility and/or purchase their ticket close to departure and to have a short permanence at destination (which does not normally include a weekend).8

15. The Commission has requested and analysed sales data from the parties and their competitors. The analysis of the parties' data indicates the existence of strong correlation between business flexible and economy flexible fares, while no meaningful correlation  is observed between business and economy restricted tickets.9 This result appears to suggest that, besides other key characteristics such as comfort of travel, flexibility is a key driving factor in determining the pattern of substitution between different types of tickets.

16. However, for the purpose of the present case, the conclusion on whether time sensitive and non-time sensitive passengers belong to the same market can be left open as no competition concerns arise under any alternative market definition.

A.3 Direct (non-stop) vs. indirect (one-stop) flights10

17. The Commission has considered in previous decisions11 that with respect to short-haul routes (below 6 hours flight duration) indirect/one-stop flights generally do not provide a competitive constraint to direct/non-stop flights absent exceptional circumstances (for example the direct flight does not allow for a one-day return trip, or the share of indirect flights in the overall market is significant). However, with the increased flight time, indirect flights constitute a more credible alternative to direct flights and their attractiveness in the eyes of passengers should be assessed on a route-by-route basis.12 This is in line with Commission's practice to consider that with respect to long-haul routes (more than 6 hours flight duration), indirect flights constitute a competitive alternative to direct services under certain conditions (for example if they are marketed as connecting flights on the O&D pair in the computer reservation system).13

18. However, for the purpose of the present case, the conclusion on whether or not direct/non-stop and indirect/one-stop flights belong to the same market can be left open as no competition concerns arise under any alternative market definition.

A.4 Airport substitutability

 19. In its practice, the Commission has found that flights from or to airports which have sufficiently overlapping catchment areas can be considered as substitutes in the eyes of passengers. Such airport substitution has often been found where several airports are located in the same city. In order to correctly capture the competitive constraint that flights from/to two different airports exert on each other, a detailed analysis taking into consideration the specific characteristics of the case at hand is necessary.14

20. The parties submit that all airports serving a particular city should be included in the relevant markets. In the present case, such airport substitutability is particularly relevant for London and Barcelona, as five different airports are located in the London urban area, i.e. Heathrow (LHR), Gatwick (LGW), City (LCY), Luton (LTN) and Stansted (STN), while three different airports are located in the Barcelona urban area, i.e. El Prat, Girona and Reus.

A.4.1 London airports

21. Previous Commission decisions suggested that a certain degree of substitutability exists between the London airports, at least for non-time sensitive passengers.15

22. As far as non-time sensitive passengers travelling between London and Madrid or Barcelona are concerned, the market investigation showed that in the present case the airports of LHR, LGW and LCY are substitutable, in particular due to the availability of means of transport connecting these airports to central London. The market  investigation further provided indications that all five London airports might be substitutable due to the existing transport infrastructure and the availability of cheaper airfares at some airports which are less conveniently located.16 However, while LHR, LGW and LCY are to be considered substitutable for the purpose of the present case, it can be left open if the airports of LTN and STN are also substitutable to LHR, LGW and LCY, as this would not have any impact on the outcome of the competitive assessment.

23. Similarly, as far as time sensitive passengers travelling between London and Madrid or Barcelona are concerned, the market investigation revealed a significant degree of substitutability between the airports of LHR, LGW and LCY. The substitutability of the other two London airports, namely LTN and STN was less clear although the market investigation gave some indications that, under certain circumstances (such as proximity of offices to this airport, etc.), even these two airports could be substitutable to the three other London airports.

24. More specifically, the analysis of the parties' data show the existence of a strong overlap and correlation between fares charged by the parties for flights from different London airports. In particular, BA appears to apply the same fares for business flexible and for economy flexible tickets – i.e. fare classes which are likely to be used by time sensitive passengers – for trips to Madrid or Barcelona regardless of which airports the plane is leaving from: LCY, LHR or LGW (see for instance Chart 1 below).

Chart 1 […]

25. The above chart reports the average revenue per coupon yielded by passengers flying BA for an outbound trip on an economy flexible ticket from London to Madrid during the period [number of years][…].17 As it can be noted, the average revenues per coupon from the three airports (LHR, LGW, and LCY) are largely overlapping and appear to evolve with approximately the same pattern throughout the whole period of observation.

26. This suggests that a certain degree of substitutability between the three airports exists, also for time sensitive passengers, as one would otherwise expect BA to discriminate passengers on the basis of their airport of origin by charging them different fares.

27. Furthermore, approximately half of all competitors and corporate customers as well as approximately two thirds of all travel agents have in the course of the market investigation underlined that they consider LHR, LGW, and LCY substitutable due to reasons such as the existing transport infrastructure between these airports and central London, and the proximity of these airports to offices of corporate customers. More specifically in relation to LGW, respondents emphasized that a sufficient level of conveniently-timed frequencies as well as cheaper airfares exist at that airport, and that the catchment areas of LGW and LHR overlap.18

28. Furthermore, the replies of a majority of both corporate customers and travel agents appear to suggest that a hypothetical monopolist in LHR would not be able to sustain a small (but not insignificant) permanent price increase for flights directed to Madrid or Barcelona because of the competitive pressure that could be exerted by potential competitors in LCY or LGW.

29. The view that LHR, LCY and LGW are sufficiently substitutable to each other in the eyes of time sensitive passengers is also shared by several of the airports concerned. In addition, the UK Civil Aviation Authority pointed out that the investigation of the UK Competition Commission into airports operated by BAA in the UK had found a significant degree of substitutability between the airports concerned.19  The UK Civil Aviation Authority further submits that such substitutability also exists for business passengers. The same conclusion was reached by the Spanish  Civil  Aviation Authority.20

30. In light of the above, for the purpose of the present case, at least the airports of LHR, LGW and LCY are considered substitutable as concerns both time sensitive and non- time sensitive passengers flying between London to Madrid and Barcelona. This result  is consistent with the findings reached in an earlier antitrust exemption decision of 2003 on a cooperation between the parties on the routes London-Madrid and London- Barcelona where LHR and LGW were found substitutable for time sensitive passengers.21

31.As concerns the substitutability of LTN and STN, this issue can be left open in the present case as no competition concerns arise under any alternative market definition.

A.4.2 Barcelona airports

32. As regards the three Barcelona airports (El Prat, Girona, Reus), which the parties consider as substitutable to each other, the market investigation widely confirmed that these airports are not substitutable at least for time sensitive passengers, mostly due to geographic distance of Girona and Reus from the Barcelona city center and the absence at Girona and Reus of a level of frequencies typically required by time sensitive passengers. In contrast, several respondents consider that non-time sensitive passengers who focus on low airfares will find these three airports substitutable to each other.22

33. However, for the purpose of the present case, the conclusion on the substitutability of various Barcelona airports can be left open as no competition concerns arise under any alternative market definition.

A.5 Inclusion of charter services 

34. The Commission has previously held that charter flights were not part of the same product market as scheduled flights.23

35. However, for the purpose of the present case, the conclusion on this issue can be left open as no competition concerns arise under any alternative market definition.24

B. Cargo

36. The Commission has found in previous cases that, with regard to air cargo transport markets, the O&D approach to market definition is inappropriate given that cargo is generally less time sensitive than passengers, and that cargo is usually transported by trans-modal means of transport "behind" and "beyond" the points of origin and destination.25 Accordingly, the relevant market should be defined more broadly.

37. The market investigation in the present case confirmed that a wider market may exist as, unlike passengers, cargo may be routed with a higher number of stopovers and therefore any indirect/one-stop route is a substitute for any direct/non-stop route.

38. With regard to cargo service operators, the market investigation confirmed that there is no reason to sub-segment the market according to the type of the provider as the four types of air cargo carriers, namely (i) cargo airlines with dedicated freighter planes; (ii) airlines with only belly space cargo capacity on passenger flights; (iii) combination airlines (i.e. airlines with both dedicated freighter airplanes and belly space cargo capacity); and (iv) integrators26, compete with each other for business with the same kinds of customers.

39. In addition, as established in previous Commission decisions, air cargo transport markets are inherently unidirectional due to differences in demand at each end of the route and must hence be assessed on a unidirectional basis.27 This was corroborated by the market investigation in this case.

40. In previous cases, the Commission left open the question whether the market for air cargo transport should be further sub-divided according to the nature of the goods transported (for example, dangerous or perishable goods).28 The parties submit that no sub-segmentation according to the type of cargo should be made. The market investigation revealed that a large majority of respondents (customers and competitors) agreed with the parties that the cargo markets should not be segmented in this way, mainly because different kinds of cargo can be transported in the same aircraft. Some respondents also pointed to the use of standardized packaging and to the fact that no different handling is required for loading and unloading. Therefore, the competitive assessment in this case will be based on the market for air cargo transport encompassing all kinds of transported goods.

41. According to previous Commission decisions and the parties' submission, the relevant market in intra-European routes of air cargo transport can be defined as European-wide and should include alternative modes of transport, notably road and train transport.29 The market investigation in the present case confirmed this view.

42. As concerns intercontinental routes, catchment areas at each end of the route broadly correspond to continents for those continents where local infrastructure is adequate to allow for onward connections (for example, by road, train, or inland waterways), such as Europe.30 In respect of continents where local infrastructure is less developed, the catchment area corresponds to the country of destination. In line with the parties' submission, the market investigation confirmed that transport infrastructure across Central America, South America and Africa is insufficient to consider these continents  as relevant catchment areas. Catchment areas in routes involving these continents must be considered on a country-by-country basis. Air cargo transport from Europe to Central America, South America or Africa (or vice versa) must thus be assessed on a continent (Europe) to country basis (the respective country in Central America, South America or Africa) (or vice versa).

43. In view of the above, the relevant cargo markets for the purpose of the present case are the ones for air transport of overall cargo on a continent-to-continent and continent-to-country basis as the case may be.

C. Groundhandling services

44. Aircraft groundhandling refers to the servicing of an aircraft while it is on the ground. It covers a variety of airport services such as ramp services (including aircraft guiding and towing, baggage loading and unloading, freight and mail handling, refuelling, de-icing, catering, lavatory drainage, water cartage, cleaning and ground power) and passenger services (including check-in counter services, gate arrival and departure services,  transfer counters, customer service and airport lounges).

45. With regard to the product market definition, the Commission considered in previous decisions that groundhandling services can be divided into several distinct segments on the basis of either the IATA Standard Groundhandling Agreement or the EU Directive on groundhandling services,31 but the determination of whether  each  segment constituted a separate relevant market was left open.32  In other cases the Commission examined possible alternative product market definitions but finally left the precise definition open.33

46. Concerning the geographic market definition, the Commission considered in previous decisions that the geographic scope for the provision (or contracting) of groundhandling services is restricted to a specific airport given that the services required at a particular airport cannot normally be substituted by services provided at other airports. The Commission thus concluded that the relevant geographic market for the provision of groundhandling services could be the area of a specific airport.34 The Commission has also considered that a wider geographic market may be defined encompassing other airports within a catchment area that could attract airlines if groundhandling service providers were present at those airports.35

47. For the purposes of this decision, the exact product and geographic market definition of groundhandling services can be left open as no competition concerns arise under any alternative market definition.

D. Maintenance, Repair and Overhaul (MRO) services

48. With regard to the product market definition, the Commission has distinguished in previous decisions four separate segments within the MRO market, namely (i) line maintenance, (ii) heavy maintenance, (iii) engine maintenance, and (iv) components maintenance.36

49. With regard to the geographic scope of the market for MRO services, the Commission has found that heavy maintenance services might be at least EEA-wide, whereas line maintenance could be local in scope37 and engine maintenance worldwide.38 With regard to MRO services for components, the Commission has considered these services to be worldwide.39 In recent decisions, the Commission left open the geographic market definition with regard to MRO services.40

50. For the purposes of this decision, the precise scope of the product and geographic  market definition for MRO can be left open as no competition concerns arise under any alternative market definition.

V. CONCEPTUAL FRAMEWORK FOR THE ASSESSMENT OF THE PRESENT TRANSACTION

51.  Prior to assessing the impact of the transaction on the relevant markets, the conceptual framework for the assessment of the transaction must be determined. In this respect, the present transaction raises two conceptual issues.

52.  The first issue concerns the treatment of IB's and BA's alliance partners for the purposes of both the determination of affected markets and the competitive assessment of the transaction.

53.  The second issue relates to the assessment of the effects of the transaction with respect  to the UK-Spain routes where both parties already co-operate with each other pre- transaction.

A.Treatment of oneworld alliance partners

54. Both BA and IB are already members of the oneworld alliance whereby they cooperate in various areas pursuant to several framework and implementing agreements.41

55. In line with the recent Commission's practice42, the parties' respective alliance partners (other than American Airlines ("AA"), as explained in the next paragraph) are not considered for the determination of affected markets. Indeed, according to the information provided by the parties, no close links are established by the merger  between a merging party and a close partner of the other merging party. However, any cooperation between the parties and an alliance partner will be taken into account qualitatively in the competitive assessment.

56. The situation is different for AA because of the "Transatlantic Joint Business Agreement" entered into by AA, BA and IB on 14 August 2008. Pursuant to this agreement, the three parties intend to cooperate on all their flights between North America (Canada, Mexico and the US) and Europe (the European Union, Norway and Switzerland), including revenue-sharing on these routes. This cooperation was investigated by the Commission under Article 101 TFEU. On 14 July 2010, following commitments offered by the parties, which it made binding for 10 years, the  Commission concluded that there were no longer grounds for action on this agreement and closed its proceedings. Due to the intensity of this cooperation, the Transatlantic Joint Business Agreement between AA, BA and IB has been considered in the present case for the purpose of determination of affected markets as in view of the Article 9 decision, it can be reasonably expected that BA, AA, and IB will put into effect a cost and revenue sharing joint venture and jointly determine fares and capacity.

B. Pre-existing cooperation between the parties on UK-Spain routes

57. Pre-merger, the parties have had various extensive cooperation agreements in place which relate to routes between the UK and Spain.

58. In 1999, the parties agreed to code-share on routes between the UK and Spain, on some other routes to and from these countries, on routes to several other countries worldwide as well as on selected routes that connect to these routes in the UK and in Spain. Furthermore, a Frequent Flyer Programme ("FFP") cooperation was set up.43

59. In 2002, the parties signed an Alliance Agreement, which was intended to form the basis for a close global cooperation.44 The aim of this agreement was to coordinate and integrate the commercial and marketing strategies and the distribution methods worldwide, and particularly on the routes between the UK and Spain. This cooperation related to enhanced code-sharing, pricing coordination, inventory and yield management, common network planning; revenue and profit sharing (passenger and cargo) and cargo activities.

60. The Commission exempted this agreement in 2003, subject to certain commitments attached to five UK-Spain overlap routes.45

61. In 2004, within the framework of the 2003 exemption decision, the parties signed a Joint Business Agreement, implementing the Alliance Agreement with a special focus on Madrid-LHR and Barcelona-LHR.

62. The Commission's 2003 exemption decision expired on 12 September 2009 and the parties have since continued to cooperate as under the Joint Business Agreement.46 The parties argue that the Commission's assessment of the compatibility of their pre-merger cooperation with Article 101 TFEU remains valid because on the relevant routes the parties are now facing stronger competitive constraints than they had in 2003, in particular as a result of the implementation of the 2003 commitments. Indeed, following the exemption decision, IB and BA released slots at both ends of the Madrid-London route  to  easyJet  and  these  commitments  slots  have  allowed  easyJet to increase its presence on the route LGW-Madrid. More specifically, it increased its daily frequencies to four on the basis of slots obtained in application of the 2003 commitments. Subsequently, easyJet was able to maintain this level of frequencies mostly on the basis of own slots.47

63. In view of this intensive cooperation, the parties argue that the merger is not likely to change the competitive landscape on UK-Spain routes and in particular on the routes London-Madrid and London-Barcelona.

64. The effects of a merger are normally assessed on the basis of the pre-merger situation, taking into account future changes to the market that can be reasonably predicted.48

65. In the present case, it is to be recalled that the parties' cooperation on the UK-Spain routes has been exempted by the Commission until very recently, i.e. September 2009. As part of this exemption decision, commitments pertaining to slots also on the route London-Madrid were enacted which have in the meantime been used by easyJet to improve its competitive position on that route, along the lines of the 2003 exemption decision, notably by becoming a credible alternative for time sensitive passengers, as described in Section VI.A49.

66. Therefore, the market situation pre-merger reflects those commitments and their effects. This needs to be taken into account, together with the continued cooperation of the parties, in the assessment of the present merger.

67. In addition, as will be shown in Section VI.A. below, the parties will post-merger be sufficiently constrained by other competitors, in particular on the routes London-Madrid and London-Barcelona.

68. In view of these elements, the effects of the proposed merger are in the present case to be compared against the pre-merger situation of cooperation between the parties.

VI. COMPETITIVE ASSESSMENT

A. Passenger scheduled air transport services

69. The present transaction gives rise to a number of horizontal overlaps that can  be  grouped in the following categories: (i) non-stop/non-stop overlaps, (ii) non-stop/one- stop overlaps, and (iii) one-stop/one-stop overlaps.50

A.1. Non-stop-non-stop short-haul overlaps: London-Madrid and London-Barcelona

70.The proposed transaction gives rise to two non-stop-non-stop overlaps between the parties, namely on the routes London-Madrid and London-Barcelona. On London- Madrid, both parties are operating direct services.51 On London-Barcelona, IB merely markets seats as a marketing carrier on BA-operated flights by way of code-sharing.

71. As described in Section IV.A above, the three London airports in Heathrow (LHR), Gatwick (LGW) and City (LCY) are substitutable with each other as far as journeys of time sensitive and non-time sensitive passengers on the routes London-Madrid and London-Barcelona are concerned. In contrast, the substitutability of the airports of El Prat, Girona and Reus can be left open for the purposes of the present case.

A.1.1 London-Madrid

72. A total of [450,000 – 500,000] O&D passengers travelled on the route LHR/LCY/LGW- Madrid in Winter 09/10, while [750,000 – 800,000] O&D passengers travelled on the route in Summer 09.

73. The parties operate the route from LHR, LGW and LCY airports with altogether 15  daily frequencies (BA 5 frequencies from LHR and 2 frequencies from LCY, IB 8 frequencies  from  LHR).52   Competitors  all  operate  from  LGW,  i.e.  easyJet  (4 daily frequencies in Winter 09/10)53, Air Europa (2 daily frequencies), and Ryanair (1 daily frequency).54

74. As regards the market encompassing the airports of LHR, LCY and LGW on the one hand and Madrid airport on the other hand, which is the relevant market for the assessment of the impact of the transaction in this case, the market position of the parties and their competitors is as follows:55

Table 1

Source for market data for all direct-direct overlaps: Form CO based on CAA56 and MIDT data

75. EasyJet is a well established player on the route connecting London to Madrid. Indeed,  it has bases57 at LGW, where it accounts for nearly a third of all slots and passengers travelling to and out of that airport, as well as bases at LTN, STN and Madrid.

76. Air Europa, a member of the Skyteam alliance, is one of the largest Spanish airlines  with a hub at Madrid airport.

77. Ryanair has not only operations at LGW but also bases at STN, LTN and Madrid  airport.

Time sensitive passengers

78. Corporate customers widely consider that under certain circumstances (in particular the availability of frequencies at convenient times) their employees would travel with low cost carriers on their business trip on London-Madrid.58 This is illustrated by the  example of a corporate customer who allows its employees to "book tickets on low- cost/chartered flights independently and reclaim the fare through [the] expenses system".59 Similarly, the parties have provided evidence on systems developed by travel agents which allow corporate customers to book business travel on low cost carriers.60 Most corporate customers and travel agents also regularly compare prices between network carriers and low cost carriers.

79. Many competitors, corporate customers and travel agents further suggested that easyJet, Air Europa and, to a lower extent, Ryanair, constrain the parties on the route London- Madrid, also for time sensitive passengers.

80. As evidenced by market data collected in the course of the market investigation, easyJet, Air Europa, and Ryanair all carry business passengers on the London-Madrid route.61

81. In relation to easyJet, respondents stressed that it offers lower fares than the parties, a considerable level of conveniently-timed frequencies, the possibility to book tickets via GDS and that it is "increasingly offering business travellers services".62 EasyJet (and Air Europa) have also often been mentioned as close alternatives to both parties that will remain  on  the  market  post  merger.63   EasyJet  itself  actively  targets  time sensitive passengers.64 In contrast, constraints stemming from Ryanair's services on London- Madrid appear mostly due to low airfares, and would therefore be less of a constraint for time sensitive passengers.65

82. Also an analysis of easyJet's data by the Commission show that, consistently with the findings above, time sensitive passengers would indeed consider easyJet as a potential alternative option when they purchase their tickets.

83. The competitive constraint exerted by easyJet is further underlined by internal documents of the parties as well as by correspondence between IB and one of IB's biggest corporate customers, who recently shifted [percentage][…] of its demand from the parties to easyJet and thus forced IB to offer it a special tariff.

84. In relation to Air Europa, the analysis of the parties' data shows that the entry of Air Europa on the route LGW-Madrid could have had a negative impact on the parties' pricing policy. In particular, a significant drop in the fares charged by IB for business flexible tickets on the LHR-Madrid is observed approximately at the same time of Air Europa's entry in LGW-Madrid (a similar drop in fares is observed on the route LHR- Barcelona). This change in IB's (and, to a smaller extent, BA's) pricing policy could indicate a reaction to the actual or the potential increase in the competitive pressure felt by the parties on the routes of concern. According to IB, such a drop in its business fares, obeys to the increased pressure imposed by low cost carriers.

85. The above findings, i.e. that a significant competitive constraint on the parties is exerted by easyJet, Air Europa, and to a lesser extent Ryanair, are confirmed by an analysis of the level and timing of frequencies of these players on the route:

86. In particular, in Winter 09/10, easyJet offered 4 daily frequencies (i.e. in the early morning, late morning, late afternoon and evening at both ends of the LGW-MAD route) which are convenient for time sensitive passengers at either end of the route. The responses of corporate customers in the present case confirmed that the number of frequencies such as offered by easyJet is sufficient to compete on the London-Madrid route for time sensitive passengers.66

87. Air Europa offers two daily flights in the early/mid morning and mid/late afternoon which are particularly attractive for passengers starting their journey in Madrid.

88. Furthermore, the market investigation revealed that corporate customers and travel agents are widely ready to use two different carriers for a same-day return trip. Time sensitive passengers thus appear able to combine the services of easyJet, Air Europa and Ryanair67 at the airports of LGW and Madrid. These players altogether offer 7 daily frequencies on the LGW-Madrid route throughout the entire day.

89. In addition to the above, easyJet's services at LTN airport (3 daily services) are particularly convenient for time sensitive passengers starting their journey in the UK, and Ryanair's services at Stansted airport (2 daily services) are particularly convenient for time sensitive passengers starting their journey in Spain. These services exert an additional competitive constraint on the parties on the route LHR/LGW/LCY-MAD. This is in line with the questionnaire replies of several corporate customers and travel agents who consider LTN and STN airports substitutable with LHR, LGW and LCY as well as with internal documents of a competitor showing that the catchment areas of LHR and LTN partly overlap.68

Non-time sensitive passengers

90. As indicated in Table 1 above, the parties would have a [30-40]% combined market share (Winter 09/10) in the non time sensitive passenger market, and would be confronted to strong competition constraints from easyJet, Air Europa, and Ryanair,  with market shares of [30-40]%, [20-30]%, and [10-20]% respectively.

91. The responses to the market investigation confirm the strong competitive constraints exerted by easyJet, Air Europa, and Ryanair, as has been widely stressed by corporate customers and travel agents.

92. Moreover, as mentioned in the preceding paragraphs, further competitive constraints emanate from easyJet's and Ryanair's operations at LTN and STN.69

Barriers to entry

93. As regards the availability of slots at the relevant airports, the market investigation showed that LHR is highly congested70 while LGW, MAD and LCY are generally congested at peak times.71 While the existing slot congestion at these airports renders entry and/or expansion more difficult, it is recalled that all players already active on the Madrid-London route have hubs or bases at least at one end of the route (easyJet at  LGW and  Madrid,  Air  Europa  and  Ryanair  at Madrid). These players thus have the potential to expand their services on the basis of their current slot holdings at LGW and Madrid. In addition, easyJet is the biggest slot holder at LGW (with nearly a third of the total slots available), and Ryanair has considerably extended its capacity at Madrid and the airport is now amongst its largest bases. Accordingly, a respondent to the market investigation has mentioned in relation to Ryanair that "their considerable slot holding  in Madrid provides the potential for greater competition" on the route.72

94. In addition, despite current capacity limitations, it is to be underlined that recent entry occurred on LGW-Madrid where both Air Europa and Ryanair started their services in 2009. It is therefore not excluded that further players already active at one end of the route might enter the route.

Conclusion

95. It follows from the above that strong competitive constraints emanate from easyJet, Air Europa and Ryanair.

96. In addition, it is recalled that the effects of the present transaction are to be compared against the pre-merger cooperation between the parties. In this scenario, the effects of the proposed merger are confined to the transformation of the parties' previous contractual cooperation into a permanent structural link.

97. On the one hand, the creation of such a permanent structural link might eliminate residual actual competition between the parties. However, the market investigation showed that due to the pre-merger cooperation between the parties, only a limited amount of residual actual competition exists between them, as they might for instance sometimes apply different terms and conditions to their corporate customers and travel agents. However, considering notably the analysis made above about the competitive strength of easyJet, Air Europa and Ryanair, the barriers to entry on the route and the degree of pre-merger cooperation (revenue and profit sharing on the route London- Madrid) between the parties, the amount of residual competition possibly eliminated by the transaction is rather low and does not give rise to competition concerns on this route.

98. On the other hand, the creation of a permanent structural link might have a detrimental effect on actual competition exerted by third parties, notably by easyJet. In this respect,  it is recalled that following the 2003 exemption decision IB and BA released slots at  both ends of the Madrid-London route to easyJet which helped easyJet to expand its presence on the route. However, while easyJet still uses one slot at Madrid airport which it obtained from the parties under the antitrust commitments of 2003, this slot is in the afternoon and can thus likely be obtained under the normal slot allocation procedure even if the parties were in a position to claim back that slot from easyJet.73 In view of the existing competition on the route, the end of the application of the commitments under the 2003 exemption decision would therefore not materially alter the competitive strength of easyJet on the route.

99. In conclusion, despite the parties' high combined market shares in particular in the time sensitive passenger market, given the existing pre-merger cooperation between the parties and the competitive strength of the parties' competitors active on that route74, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

A.1.2 London-Barcelona

100. A total of [250,000 – 300,000] O&D passengers travelled on the route LHR/LCY/LGW-Barcelona in Winter 09/10, while [550,000 – 600,000] O&D passengers travelled on the route in Summer 09. These seasonal fluctuations, whereby twice as many passengers travel on this route in the summer season, illustrate that many passengers on this route travel for leisure purposes.

101. BA operates the route with 8 daily frequencies from LHR and LCY to El Prat while IB stopped operating as of Winter 09/10 and is now marketing seats on BA-operated flights by way of code-sharing.75 The other airline offering competing direct services on this route is easyJet with 3 daily frequencies out of LGW.76

102. As regards the market encompassing the airports of LHR, LCY and LGW on the one hand and El Prat airport on the other hand, which is the relevant market for the assessment of the impact of the transaction in this case, the market position of the parties and their competitors is as follows:77

Table 2

103. EasyJet is a well established player on the route connecting London to Barcelona. It has bases at LGW, where it accounts for nearly a third of all slots and passengers travelling to and out of that airport, as well as LTN and STN. Furthermore, easyJet is reported to currently negotiate the opening of a base at El Prat.78

Time sensitive passengers

104. As in the case of the route London-Madrid, time sensitive passengers appear to consider easyJet as a viable competitive alternative (see above Section VI.A.1.1). This is confirmed by easyJet's considerable market share in the time sensitive passenger market and its conveniently-timed daily frequencies, i.e. at the London end in the early  morning, afternoon and evening (and during the current Summer season in addition also in the late morning), and at the Barcelona end in the late morning, late afternoon and  late evening (and during the current Summer season in addition also in the evening). Furthermore, a study of the UK CAA of May 2009 found that on the route London- Barcelona, easyJet's LGW service attracted a higher proportion of business passengers than both BA's LGW and LHR services.79

105. In line with the above, many competitors, corporate customers and travel agents suggested in the course of the market investigation that easyJet constrains the parties on the route London-Barcelona, also for time sensitive passengers.

106. In addition, some competitive constraints also appear to emanate from easyJet's services at LTN (2 daily frequencies) and STN (1 daily frequency), as evidenced in particular by the questionnaire replies of several travel agents and the UK CAA of May 2009.80

Non-time sensitive passengers

107. As indicated in Table 2 above, the parties would have a [50-60]% combined market share (Winter 09/10) in the non-time sensitive passenger market, and would be confronted to strong competition constraint from easyJet with a market share of [40- 50]%.

108. The responses to the market investigation confirm the strong competitive constraints exerted by easyJet, as has been widely stressed by corporate customers  and  travel agents.

109. In addition, Ryanair operates between LTN, LGW and STN and Girona, as well as between STN, LTN and Reus. These services exert an additional competitive constraint on the parties on the route LHR/LGW/LCY-El Prat as far as non-time sensitive passengers are concerned.81

Barriers to entry

110. As regards the availability of slots at the relevant airports, the market investigation showed that LHR is highly congested throughout the day, while LGW, LCY, and El Prat are congested at peak times in the morning and in the evening.

111. Moreover, easyJet has a base at LGW where it is the biggest slot holder with nearly  a third of all slots. Furthermore, easyJet is reported to negotiate the opening of a base at El Prat. Similarly, Ryanair (which already operates between STN/LTN/LGW and Girona/Reus since 2003) has operations at LGW and will open a base at El Prat in September 2010. Both airlines are thus candidates for expansion or entry on the route, should the parties attempt to increase prices. Further airlines with a hub/base at El Prat include Star Alliance carrier Spanair, which is a natural candidate for entry on the route due to its activity at Barcelona.

Conclusion 

112. It follows from the above that strong competitive constraints emanate from easyJet on this route.

113. In addition, it is recalled that the effects of the present transaction are to be  compared against the pre-merger cooperation between the parties. In this scenario, the effects of the proposed merger are confined to the transformation of the parties' previous contractual cooperation into a permanent structural link.

114. The creation of such a permanent structural link might eliminate residual actual competition between the parties. However, similar to the findings in Section VI.A.1.1 above, the elimination of even actual competition between the parties cannot be considered sufficiently significant so as to lead to competition concerns on this route, notably due to the degree of pre-merger cooperation, the competitive strength of easyJet and the barriers to entry on the route.

115. Furthermore, as the situation outlined in the preceding paragraph will in any event sufficiently restrain the parties post-merger, the transaction does not give rise to competition concerns by a possible elimination of IB (or its subsidiary Vueling) as a most likely future potential competitor on the route.82

116. In conclusion, despite the parties' high combined market shares in particular in the time sensitive passenger market, given the pre-merger cooperation between the parties and the competitive strength of easyJet83, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route,.

A.2. Non-stop/one-stop overlaps

117. The present transaction gives rise to a number of horizontal overlaps on short-haul and long-haul routes where one party offers a non-stop connection while the other party offers a one-stop connection. Some of these routes have one end outside the European Union and have less than 30,000 passengers. Therefore, they are too thin to be considered as a substantial part of the internal market. The Commission has therefore limited its investigation to the overlaps on the routes analysed below.

A.2.1. Non-stop/one-stop overlaps on short-haul routes

118. As indicated above in section IV.A.3 on market definition, one-stop services generally do not constitute a competitive constraint on non-stop services as far as short- haul routes are concerned. However for a number of short-haul routes concerned by the present transaction, a significant proportion of passengers use one-stop services. This fact could indicate that customers consider these one-stop flights as alternative to non- stop flights.

119. However, in the present case it can be left open whether non-stop and one-stop flights are substitutable on the routes analysed below as it would not change the conclusion of the competitive assessment. If non-stop and one-stop flights were considered as two separate markets there would be no overlap between IB and BA. If non-stop and one-stop flights were in the same market, the transaction would lead to an overlap but would not give rise to competition concerns on any of these routes for the reasons explained further below. This analysis is valid for time sensitive and non-time sensitive passengers.

Non-stop/one-stop overlaps on intra-European short-haul routes

120. The parties’ activities overlap on a number of short-haul routes between the UK and Spain (i.e. London-Ibiza, London-Malaga, London-Palma and  London-Alicante), Ireland and Spain (Barcelona-Dublin)and between the Netherlands and Spain (Amsterdam-Malaga), where one party offers non-stop services and the other party  offers one-stop services. These routes can be deemed not to raise competition concerns, even in case non-stop and one-stop flights were substitutes84, if account is taken of the sales of easyJet and Ryanair on the UK-Spain routes85, of the sales of Aer Lingus on Barcelona-Dublin and of those of Transavia on Amsterdam-Malaga. All these carriers provide non-stop services. Their sales account for a very large share of total sales made on these routes. On all routes at least one of these players achieves market shares of  more than 40%, which is significantly higher than the parties' combined market share.

121. The activities of the parties result in two additional non-stop/one-stop overlaps on short-haul intra European routes: Barcelona-Edinburg and Madrid-Dubrovnik. On these routes, the combined market share of the parties is significant.

122. As regards the Barcelona-Edinburgh route, BA provides one-stop services while IB's subsidiary, Vueling, started operating this route with non-stop flights in June 2010 for the peak months of July and August. The number of passengers on this route, including all sales, was [20,000 – 30,000] in 2009. The percentage of passengers flying one-stop was approximately [80-90]% in Summer 09 and [10-20]% in Winter 09/10. The combined market share of the parties on this route was [50-60]% in Summer 09 and [40- 50]% in Winter 09/10. At least two credible competitors will remain post-merger, Air France/KLM86 and Lufthansa. A possible inclusion of Girona and Reus airports into the relevant market does not have an effect on the conclusion of the competition assessment.

123.Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

124. The Madrid-Dubrovnik route involves a non-stop/one-stop overlap between IB's non-stop services and BA's one-stop services. The number of passengers on this route in 2009 was [10,000 – 20,000] in Summer 09 and [0-5,000] in Winter 09/10. The percentage of passengers flying one-stop was approximately [5-10]% in Summer 09 and [10-20]% in Winter 09/10.

125. If the charter airline Dubrovnik Airlines, who operates non-stop services on the route, were included in the relevant market, the parties' combined market share would be [40-50]% in Summer 09 and [0-5]% in Winter 09/10 while Dubrovnik Airlines' market share (non-stop services) would be [30-40]% in Summer 09, and [80-90]% in Winter 09/10.87

126. If Dubrovnik Airlines were not included in the relevant market, the parties'  combined market share would very high (i.e. around [90-100]% in Summer 09, while  the market share would be below [10-20]% in Winter 09/10). However, the increment brought about by the transaction is small (i.e. BA had a market share of [5-10]% in Summer 09), and further carriers such as Lufthansa/Swiss (market share of [5-10]% in Summer 09) operate on the route, while Spanair recently entered the market and is thus likely to replicate the constraining effect that BA had previously on IB.

127. Therefore, although one respondent to the market investigation has expressed concerns about the effects of the transaction on this route, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route, irrespective of whether or not Dubrovnik Airlines is included in the relevant market.

A.2.2. Non-stop/one-stop overlaps on long-haul routes

128. As indicated in the market definition section, with regard to long-haul routes, one-stop flights constitute a competitive alternative to non-stop services under certain conditions. Substitutability between non-stop and one-stop services shall however still be assessed on a route-by-route basis. Notably, a significant proportion of passengers use one-stop services. This very fact could indicate that customers consider these one-stop flights as alternative to non-stop flights. The parties are also of the opinion that one-stop services provide credible alternatives to non-stop services in the routes analysed below.

129. Notwithstanding the above, it can be left open whether non-stop and one-stop flights are part of the same market for the routes analysed below as it would not change the conclusion of the competitive assessment.

(1)  UK-South America routes

London-Rio de Janeiro ([60,000 – 70,000] passengers in 2009)

130. The London-Rio route is a route with [60,000 – 70,000] O&D passengers in 2008 and [60,000 – 70,000] passengers in 2009. BA offers non-stop services on this route  with 3 weekly frequencies while IB offers one-stop services via Madrid. More than [50- 60]% of the passengers on this route choose to fly one-stop.

131.   The table below shows the market shares of the parties and their main competitors

Source: Form CO-based on MIDT data

132. Although the parties' combined market share on this route is high, the merged entity will face significant competition from TAM's non-stop services. TAM has a base at Rio de Janeiro's airport (GIG) and serves this route with 7 weekly frequencies. Air France and TAP are also credible competitors.

133. The market investigation has not revealed any substantiated concerns about the impact of the transaction on this route, although one participant has pointed to the high combined market shares of the parties. In spite of these concerns and for the reasons mentioned in the precedent paragraph, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

London-Sao Paulo ([100,000 – 150,000]  passengers in 2009)

134. The London-Sao Paulo route is a route with [100,000 – 150,000] O&D passengers in 2008 and [100,000 – 150,000] passengers in 2009. BA offers non-stop services on this route with 7 weekly frequencies while IB offers one-stop services via Madrid. Almost [50-60]% of passengers on this route choose to fly one-stop.

135.   The table below shows the market shares of the parties and their main competitors.

Source: Form CO-based on MIDT data

136. The parties' combined market share is high in the market for time sensitive passengers in the Summer IATA season 09 ([60-70]%) but lower in Winter 09/10 ([50- 60]%). There are 5 other alternative credible competitors, two of which achieve significant market shares in this market (TAM, [10-20]% in Summer 09 and [10-20]%  in Winter 09/10 and TAP, [0-5]% in Summer 09 and [10-20]% in Winter 09/10). TAM has the same number of frequencies on this route as BA (7 frequencies) and a base at  Sao Paulo's airport (GRU).

137. The market investigation has not revealed any substantiated concerns about the impact of the transaction on this route, although one participant has pointed to the high combined market shares of the parties. In spite of these concerns and for the reasons mentioned in the precedent paragraph, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

(2)  UK-South Africa routes

London-Johannesburg ([350,000 – 400,000]  passengers in 2009)

138. London-Johannesburg is a route with [350,000 – 400,000] O&D passengers in 2008 and [350,000 – 400,000] passengers in 2009. BA offers non-stop services on this route with 19 weekly frequencies in Summer and 14 weekly frequencies in Winter. IB offers one-stop services via Madrid. More than [30-40]% of the passengers on this O&D choose to fly one-stop.

139. The parties compete on this route with two other airlines offering non-stop services (South African and Virgin Atlantic) and 6 airlines offering one-stop services.

140. The table below provides market share information in the last 2 IATA seasons.

Source: Form CO-based on MIDT data

141. The combined market share of the parties for all passengers remains below [30-40]% and the increment brought about by the merger is of less than [0-5]% in all possible markets. In the market for time sensitive passengers, the parties' combined market share is of [40-50]% in Summer 09 and [40-50]% in Winter 09/10. The merged entity will face competition from two main carriers providing non-stop services on this route, Virgin Atlantic and South African, with considerable market shares in the market for time sensitive passengers (Virgin, [20-30]% in Summer 09 and [20-30]% in Winter 09/10; South African, [10-20]% in Summer 09 and [10-20]% in Winter 09/10). Virgin Atlantic has a base at London airport (LHR) and serves this route with 7 weekly frequencies. South African has a base at Johannesburg airport (JNB) and operates this route with 14 weekly frequencies. In addition, the merged entity will be constrained by 6 other carriers providing one-stop services (Emirates, Air France, Qatar, Lufthansa/Swiss, Egyptair, Etihad).

142. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

(3)  UK/Spain-North America routes

143. American Airlines is present on the UK/Spain-North American routes where the activities of the parties overlap. Having regard to the Transatlantic Joint Business Agreement between the parties and this carrier, and given the adoption by the Commission of an Article 9 decision on 14 July 2010, the assessment of these routes will focus on the merger-specific effects of the transaction, taking into consideration the close cooperation that will exist between the merged entity and AA as a result of the mentioned Alliance.

Mexico routes

London-Mexico City ([50,000 – 60,000] passengers in 2009)

144. The London-Mexico City route is a route with [60,000 – 70,000] O&D passengers in 2008 and [50,000 – 60,000] passengers in 2009. BA offers non-stop services on this route with 4 weekly frequencies in Summer and 3 weekly frequencies in Winter, while IB offers one-stop services via Madrid. As the parties indicate, almost [40-50]% of all passengers on this route fly one-stop.

145. Since January 2009, Mexicana offers non-stop services on this route with 4 weekly frequencies in Summer and Winter out of LGW. The parties' services are provided out  of LHR. Mexicana has achieved on a short-time frame a significant market share. It has a hub at Mexico City's airport (MEX). The parties are of the view that Mexicana's recent entry evidences the lack of barriers to enter at either end of the London-Mexico City route. In addition, there are 6 carriers that provide one-stop services out of LHR.

146. The parties' market shares and those of their main competitors are shown in the table below.

Source: Form CO-based on MIDT data

147. In spite of the high combined market shares of the parties and American Airlines, particularly on the market for time sensitive passengers, the increment brought by the merging parties as regards this market is of [0-5]% in the IATA Season Summer 09 and [0-5]% in the IATA season Winter 09/10. Mexicana is particularly strong in this time sensitive passengers market with market shares of [20-30]% in the IATA Season Summer 09 and [30-40]% in the IATA Season Winter 09/10. The merged entity would not only be constrained by Mexicana's non-stop services as regards all types of passengers but also by the one-stop services provided by Air France/KLM, Continental and other competitors (Lufthansa, Delta and Aeromexico).

148. A few competitors representing a minority of the participants in the market investigation have expressed concerns about the impact of the transaction on this route. Only the submission of one of these competitors is substantiated. In its opinion, Mexicana's operations will not constrain the parties' activities because BA and  Mexicana have entered into a code-share agreement on the London-Mexico route and, in addition, Mexicana has recently joined the oneworld alliance. The combined oneworld members market share, consisting of BA, IB, AA and Mexicana would be about [80- 90]% on the LGW and LHR-Mexico City routes, and BA/IB would have a combined market share of about [60-70]%. On LHR-Mexico City, the share of BA/IB would be of about [80-90]%.

149. The existence of a code-share agreement between BA and Mexicana on the London- Mexico City route has been denied by the parties and the market investigation has not provided any further arguments contradicting the parties' statements on this issue. As regards the relevance of Mexicana's oneworld membership, although such membership is considered, it does not allow to add the market shares of a third party (i.e. those of Mexicana) to those of the merging parties, unless there were additional factors leading  to consider that the merging parties and the relevant third party effectively operate as a single entity. Any such factor is absent in this case. As for the presence of AA on this route, its market share is not high ([5-10]% in the worst case scenario and less than [0-5]% in the time sensitive passengers' market) and does not add much to the parties' market share. Finally, it can be considered that Mexicana's non-stop flights provide a sufficient competitive constraint to IB/BA operations.In addition, even if Mexicana was left aside, the services on this route of Air France/KLM, Continental and other smaller competitors providing services out of LHR would still provide alternatives to the services of the merging parties and AA for all passenger categories.

150. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

Madrid-Mexico City ([100,000 – 150,000] passengers in 2009)

151. The Madrid-Mexico City route is a route with [150,000 – 200,000] O&D passengers in 2008 and [100,000 – 150,000] passengers in 2009. IB offers non-stop services on this route with 7 weekly frequencies in Summer and in Winter while BA offers one-stop services via London. According to the parties, approximately [10-20]% of all passengers on this route fly one-stop.

Source: Form CO-based on MIDT data

152. The combined market share of the parties on this route remains below [30-40]%, even in the market for time sensitive passengers. The increment brought by the merging parties is less than [5-10]% for all types of passengers. A few participants in the market investigation have shown concerned about the impact of the transaction on this route without providing further details. However, the Commission is of the view that the merged entity will face strong competition from Aeromexico's and Mexicana's non-stop services in all market segments. Aeromexico is the leading player in the market for non- time sensitive passengers. Mexicana entered this route in February 2009  and immediately lead the market for time sensitive passengers.88  Both companies have a base at Mexico City (MEX) airport. In addition there is a third player providing one-stop and non-stop services, Air Europa, as well as four competitors providing one-stop services.

153. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

US routes

154. Two participants in the market investigation have expressed concerns (not substantiated) as regards the effects of the transaction on these routes, with the exception of the Madrid-New York route.

Madrid-Boston ([30,000 – 40,000] passengers in 2009)

155. The Madrid-Boston route is a route with [30,000 – 40,000] O&D passengers in 2008 and [30,000 – 40,000] passengers in 2009. IB offers non-stop services with 7 weekly frequencies in Summer and 3 weekly frequencies in Winter. BA offers one-stop services via London. More than [50-60]% of the passengers on this route choose to fly non-stop.

156. IB and American Airlines have a code-share agreement ([description of the code-share agreement][…]) to be replaced by the above mentioned Transatlantic Joint Business Agreement with AA which will imply revenue-sharing.

157. Apart from the parties and AA, there are several carriers offering one-stop services on this route. Their market shares are shown in the table below.

Source: Form CO-based on MIDT data

158. The parties' market share on this route is high, particularly in the market for time sensitive passengers (together with AA, [70-80]% in the IATA season, Summer 09). The overlap is however less than [10-20]%. Post merger, the merging parties will still face significant competition in this market from Lufthansa/Swiss ([10-20]% in Summer 09 and [10-20]% in Winter 09/10) and US Air ([0-5]% in Summer 09 but [10-20]% in Winter 09/10). Air France/KLM and Delta are credible competitors in the market for non-time sensitive passengers. There are 4 other carriers providing one-stop services on these routes. Therefore, post-merger, sufficient credible alternatives remain for all passengers markets. In addition, US Airways (member of Star Alliance) and Delta (member of SkyTeam) have significant operations at Boston airport (BOS). As a result they are particularly well placed to offer non-stop services either alone or together with their Madrid-based partners (Spanair and Air Europa, respectively). Air Europa is expanding in the US: it has recently entered the Madrid-Miami and Madrid-New York routes.

159. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

Madrid-Chicago ([30,000 – 40,000] passengers in 2009)

160. The Madrid-Chicago route is a route with [30,000 – 40,000] O&D passengers in 2008 and [30,000 – 40,000] passengers in 2009. More than [40-50]% of the passengers on this route choose to fly one-stop. IB provides non-stop services with 7 weekly frequencies in Summer and 6 in Winter whereas BA provides one-stop services via London.

161. IB and American Airlines have a code-share agreement ([description of the code-share agreement][…]), to be also replaced by the Transatlantic Joint Business Agreement  between the parties and this carrier.

162. Apart from IB and American Airlines, there are no other carriers providing non-stop services. There are however 9 other carriers competing on this route with one-stop services. Their market shares are shown in the table below.

Source: Form CO-based on MIDT data

163. The combined market share of the parties together with AA's is high in all market segments ([70-80]% in the worst case scenario). However, in view of its current market share, Lufthansa/Swiss will be a significant competitor of the merged entity and AA in the market for time sensitive passengers, with market shares of [10-20] % in Summer 09 and [10-20]% in Winter 09/10. In this latter market, the increment brought by the merging parties is of only [0-5]% in the last IATA season. US Air and United also have a non- negligible market share in this market. Concerning non-time sensitive passengers, the main alternatives would be Delta ([5-10]% in Summer 09 and [5-10]% in Winter 09/10) and US Air ([5-10]% in Summer 09 and [5-10]% in Winter 09/10). In addition there are 6 other smaller competitors providing one-stop services.

164. Madrid airport, even though there are some capacity constraints at peak hours, has available capacity for such long-haul flights. While Chicago airport is constrained, United has a hub there and would be well-placed to offer non-stop services either alone of together with its Madrid-based alliance partner, Spanair. As discussed above, Air Europa is also expanding in the US and could therefore become a new entrant on this route should the parties increase prices as a result of this merger.

165. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

Madrid-Miami ([80,000 – 90,000] passengers in 2009)

166. The Madrid-Miami route is a route with [80,000 – 90,000] O&D passengers in 2008 and [80,000 – 90,000] passengers in 2009. Approximately [10-20]% of the passengers on this route choose to fly one-stop. IB provides non-stop services with 7 weekly frequencies in Summer and Winter whereas BA provides one-stop services via London. IB code-shares on this route with AA. The code-share agreement will be replaced by the Transatlantic Joint Business Agreement implying revenue-sharing. AA provides non-stop services with 7 weekly frequencies in the Summer and the Winter seasons.

167. The parties' market shares and those of their competitors are shown in the table below.

Source: Form CO-based on MIDT data

168. The parties have high combined market shares on this route reaching in the worst case scenario [50-60]%. AA's market share is also high ([40-50]% in the worst case scenario). US Airways and Air Europa will be the main competitors of the parties that will remain post-merger. US Airways provides one-stop services and achieves a market share of approximately [5-10]%. Air Europa's market share is of less than [5-10]%. However, this carrier, which has been providing one-stop services in the past, launched a new non-stop service in March 2010 and it is thus likely to increase its market share. As indicated by the parties, Miami airport (MIA) is not congested and Air Europa's recent launch of non-stop services evidences the lack of entry barriers on this route.

169. In addition, the increment brought by the merging parties is less than [0-5]% irrespective of the market considered. Finally, IB, BA and AA have offered commitments on this route further to the "Transatlantic Joint Business Agreement" in relation to which the Commission adopted a decision on 14 July 2010. These commitments are aimed at facilitating further entry / expansion on the route.

170. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

Madrid-New York ([250,000 – 300,000] passengers in 2009)

171. IB offers non-stop services on this route with [250,000 – 300,000] passengers in 2008 and [250,000 – 300,000] passengers in 2009. BA offers one-stop services. AA is present on this route. The existing code-share agreement between AA and IB will be replaced by the above mentioned Transatlantic Joint Business Agreement. Approximately [10-20]% of the passengers on this route choose to fly one-stop.

Source: Form CO-based on MIDT data- (*)Delta code-shares with Air France/KLM

172. The market share of the parties and AA on this route remains below [30-40]% irrespective of the passenger market considered. The increment brought by the merger is less than [0-5]%. In addition, there are four other carriers providing non-stop services on this route, three of which achieve considerable market shares (Air Europa, Delta and Continental), along with 5 other carriers providing one-stop services.

173. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

Madrid-Washington DC

174. IB offers non-stop services on this route with [30,000 – 40,000] passengers in 2008 and [30,000 – 40,000] passengers in 2009. BA offers one-stop services. Approximately [60- 70]% of the passengers on this route choose to fly one-stop. IB and AA have a code-share agreement ([description of the code-share agreement][…]), to be replaced by the Transatlantic Joint Business Agreement.

175. United Airlines and Aer Lingus have recently started to operate a joint transatlantic non-stop service between Washington and Madrid. The route is served by 7 other carriers providing one-stop services.

176. The market shares of the carriers present on this route are shown in the table below.

Source: Form CO-based on MIDT data- (*)Delta codeshares with Air France/KLM

177.The combined market share of the parties and AA in Summer 09 was approximately [60-70]% in the market of time sensitive passengers and [60-70]% in the market of non- time sensitive passengers. Their market shares went down to approximately [30-40]% in all segments in Winter 09/10, when United Airlines and Aer Lingus launched their joint transatlantic non-stop service between Washington and Madrid. As indicated by the parties, during the current Summer season, IB's frequency share is [20-30]% whilst United/Aer Lingus's frequency share is [70-80]%. Furthermore, IB has decided to cancel Winter services on this route due to low profitability, whereas United/Aer Lingus will hold a [60- 70]% frequency share. There are seven carriers providing one-stop services on this route of which at least 4 can be considered credible competitors and, as indicated by the parties, the Washington airport (IAD) is not constrained and, with the exception of some capacity constraints at peak hours, there is space available at Madrid airport.

178.Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

A.3. One-stop/-one-stop overlaps

179. The activities of the parties also overlap on short-haul and long-haul routes where both operate one-stop flights.

A.3.1. One-stop/one-stop overlaps on short-haul routes

180. The short-haul routes involving one-stop/one-stop overlaps between the parties are Malaga-Budapest and Malaga-Geneva.

181. On the route Malaga-Budapest, the combined market share of the parties in the last two IATA seasons remains below [20-30]%, the increment brought about by the transaction would be small (less than [0-5]%) and passengers would have a choice amongst a number  of other competitors after the merger. In addition, BA is not actually active on Malaga – Budapest due to the poor connectivity offered over London.

182. On the route Malaga-Geneva, the combined market shares of the parties are also of less than [20-30]% and the increment brought by the merger is less than [0-5]%.

183. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to these routes.

A.3.2. One-stop/one-stop overlaps on long-haul routes

London-Buenos Aires ([80,000 – 90,000] passengers in 2009)

184. The London-Buenos Aires route is a route with [70,000 – 80,000] O&D passengers in 2008 and [80,000 – 90,000] passengers in 2009. BA's service is a one-stop service via Sao Paulo marketed as a direct flight with 7 weekly frequencies in Summer and in Winter. IB offers one-stop services via Madrid. According to the information in the Form CO, approximately [60-70]% of all passengers on this route fly one-stop.

185. Passengers on this route might choose amongst several one-stop alternative services. The market shares of the parties and those of their main competitors are shown in the table below.

Source: Form CO-based on MIDT data

186. The parties' combined market share for all passengers is [50-60]% for the IATA season Summer  09 and  [50-60]%  for  the  IATA  season  Winter  09/10.  On the  market  of time sensitive passengers, it is of [50-60]% in Summer 09 and [60-70]% in Winter 09/10. The merged entity will however face competition from a significant competitor, TAM90. TAM's market share for all passengers was [10-20]% in the IATA season Summer 09 and [10- 20]% in the IATA season Winter 09/10. In the market for time sensitive passengers, TAM's market share has decreased in the last IATA season (from [20-30]% to [10-20]%). Such decrease was compensated by the reinforcement of TAM's position in the market for non- time sensitive passengers. Overall, taking into consideration the last 4 IATA season, TAM's position has been steadily reinforced. Its market share for all passengers has increased from [10-20]% in Summer 08 to [10-20]% in Winter 09/10. Air Europa (a new entrant on this route) and Air France, together with other less significant competitors on this route (Alitalia and Lufthansa), provide further alternatives for all types of passengers. Both companies have increasing market shares in the market for time-sensitive passengers. In addition, the arguments of the parties as regards possible new entries on this route have been partly confirmed by the market investigation.

187. The market investigation has not revealed any substantiated concerns about the impact of the transaction on this route, although one participant has pointed to the high combined market shares of the parties. In spite of these concerns and for the reasons mentioned in the precedent paragraph, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to this route.

Other one-stop/one-stop overlaps on long-haul routes

188. The parties' activities overlap on a number of long-haul routes where both operate one-stop flights only. On some of the routes, AA is also present. With the exception of Barcelona-Miami and Malaga-New York, the combined market share of the merging parties does not exceed [30-40]% in any possible market. In addition, taking into account the Transatlantic Joint Business Agreement, the merged entity will still face competition from significant competitors.

189. On Barcelona-Miami and Malaga-New York, the combined market share of the parties is higher, particularly for time sensitive passengers on the route Barcelona- Miami in the Winter 09/10 season ([40-50]%), and for non-time sensitive passengers on the route Malaga-New York in Winter 09/10 ([50-60]%). One of the respondents to the market investigation has expressed not substantiated concerns as regards the effects of the transaction on the Barcelona-Miami route. However, post-merger, there will still be many competitors (apart from the parties and AA) offering one-stop services on both routes, some  with bases or hubs at one end of the route. In addition, these are routes  with a limited number of passengers that barely exceeds (or has exceeded in previous years) 30,000 passengers in 2009. Delta entered the route Barcelona-Miami, on May 2010 and the route Malaga-New York in Summer 08 IATA season.

190. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to these routes.

A.3.3. One-stop/one-stop affected routes that arise as a result of code-sharing only

191. The combination of the activities of the merging parties results in a number of affected routes that arise as a result of code-sharing only.

192. The combined market share of the merging parties does not exceed [30-40]% irrespective of the passenger segment considered in the routes Madrid-San Francisco  and Madrid-Los Angeles. In the later route mentioned, although AA is present, its market share is also moderated ([10-20]% in the worst case scenario, non-time sensitive passengers Winter 09/10) and there are strong competitors, notably Lufthansa/Swiss  with [30-40]% of the demand of time sensitive passengers in Summer 09 and [20-30]% in Winter 09/10, as well as Air France/KLM and US Air with market shares between [5- 10]% and [10-20]% depending on the IATA season and the type of passengers.

193. On London-Santiago and London-Lima, there are no non-stop services. A minority of respondents to the market investigation have expressed non substantiated concerns about the effects of the transaction on these routes. However, it shall be noted that the number of passengers on both routes is limited ([30,000 – 40,000] on London-Santiago and [30,000 – 40,000] on London-Lima). The combined market share of the parties in the worst case scenario was [40-50]% on London-Santiago (non-time sensitive passengers, Summer 09 and Winter 09/10) and [50-60]% on London-Lima (non-time sensitive passengers, Winter 09/10) and, strong competitors offering one-stop services will remain post-merger.

194. More precisely, on London-Santiago the competitors of the merged entity would be LAN Chile with [20-30]%, Air France/KLM with [10-20]% and TAM with [10-20]% of the demand of time sensitive passengers in Winter 09/10. On London-Lima, TAM with  a [50-60]% market share in the time sensitive passengers market in Summer 09 as well as Air France with [5-10]% and Continental with [0-5]% of the demand of time sensitive passengers respectively. In Winter 09/10 TAM carried [10-20]% and Air France [10- 20]% of the demand of time sensitive passengers respectively.

195. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to these routes.

A.3.4. Additional affected routes due to aggregation of American Airlines' market shares on transatlantic routes.

196. The combination of IB/BA together with AA results in high market shares in a number of routes where the merging parties' combined market share does not exceed [20-30]%, irrespective of the market segment considered. Some of these routes have one end outside the European Union and have less than 30,000 passengers. Therefore, they are too thin to be considered as a substantial part of the internal market. The Commission has therefore limited its investigation to the following routes: Barcelona-New York, Rome- Chicago, Paris-Miami and Milan-New York.

197. All these routes involve only one-stop/one-stop overlaps between the merging parties. The overlap brought by the merger is less than [5-10]%. In addition, on all these routes, significant competitors will remain post-merger, apart from the parties and AA. More precisely on the Barcelona-New York route, the merged entity and American Airlines will compete with Delta ([20-30]% in Summer 09 and [30-40]% in Winter 09/10, all passengers) and Continental ([10-20]% in Summer 09 and [10-20]% in Winter 09/10, all passengers) and with other players, including Air France/KLM. On the route Rome-Chicago, the main competitor of the merged entity and AA will be Alitalia ([30- 40]% in Summer 09 and [40-50]% in Winter 09/10, all passengers). On the route Paris- Miami, the main competitive constraint will result from the presence on this route of Air France ([20-30]% in Summer 09 and [30-40]% in Winter 09/10, all passengers) and Delta ([10-20]% in Summer 09 and [5-10]% in Winter 09/10, all passengers). On the route Milan-New York Alitalia ([10-20]% in Summer 09 and [20-30]% in Winter 09/10, all passengers), Delta ([10-20]% in Summer 09 and [10-20]% in Winter 09/10, all passengers) and Continental ([10-20]% in Summer 09 and [10-20]% in Winter 09/10, all passengers) will be present. These carriers are significant competitors on these routes also for the markets of time sensitive and non- time sensitive passengers.

198. Therefore, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to these routes.

A.3.5. Europe – US/Latin America 

199. Some participants in the market investigation have expressed concerns about the effects of the proposed transaction in a hypothetical overall market comprising all the O&D routes from Europe to the United States and Latin American countries.

200. In this respect, it is noteworthy to mention that although the merged entity will be a leading player in connecting Europe with the mentioned regions, it will compete with other strong players and its market share will remain below [20-30]%. More precisely, according to internal documents of the parties, the merged entity will have a [20-30]% share of the total capacity (average seat kilometre - ASK) existing on routes between Europe and Latin America91 and it will compete with the following main players: Air France/KLM ([10-20]%), TAM ([5-10]%), Lufthansa Group ([5-10]%) and TAP ([5- 10]%). On the routes between Europe and the US the merged entity would have a share of total existing capacity of [10-20]%. The share of American Airlines would be [5- 10]%. Other strong competitors on these routes are Delta ([10-20]%), the Lufthansa Group ([10-20]%) and Air France/KLM ([10-20]%).

201. Therefore, should the proposed geographic delimitation be of relevance to the assessment of the present case, the above figures would allow to exclude competition concerns.

B. Air transport of cargo

202. British Airways World Cargo is BA's cargo airline carrying freight, mail and courier traffic. For the air transport of cargo, BA uses belly-hold space in its passenger aircraft  as well as three full freighter aircraft.

203. Iberia Cargo is IB's cargo division. For the air transport of cargo, IB relies on capacity in the belly-hold of its passenger aircraft and on two full freighter aircraft based at Madrid airport and exclusively dedicated to intra-European traffic.

204. As regards intra-European routes of air cargo transport, the parties estimate that their combined market share in 2009 was [10-20]% calculated on the basis of a market definition of intra-European routes excluding alternative means of transport such as road and train. Given that the market investigation confirmed that the relevant market of intra-European routes of air cargo transport can be defined as European-wide and should include alternative modes of transport, notably road and train transport, and that the parties' combined market share on a narrower definition of this market does not lead to an affected market, the Commission considers that the transaction does not raise competition concerns on the market for intra-European routes of air cargo transport

205. On inter-continental routes and according to the parties, the transaction gives rise to 17 affected markets with respect to air cargo transport. The market shares of the parties on the affected markets in 2009 were as follows:

206. As can be seen from the table above, in three of the affected routes, BA's market share was below [0-5]%. In three other routes, BA's market share was below [0-5]%.

The Commission thus considers that the transaction raises no competition concerns in the six routes from Europe to Argentina, Colombia, El Salvador, Honduras, Nicaragua and Panama.

207. On three routes to Europe from Costa Rica, Dominican Republic and Guatemala, the combined market share of the parties exceeds [50-60]%. In the remainder of the routes (all of them to Europe except for the route Europe to Venezuela), the combined market share of the parties is below [30-40]% (or slightly above on the route Argentina to Europe). However, on each of these routes, there are one or more credible competitors capable of imposing a significant competitive constraint on the parties.

208. In particular, and with regard to the three routes where the parties have a market share exceeding 50%, the following credible competitors are present:

• Costa Rica-Europe: American Airlines: [20-30]%; KLM-Royal Dutch: [10- 20]%; Martinair Holland: [5-10]%; Cargolux: [5-10]%; Livingston: [0-5]%;
• Dominican Republic-Europe: Air France: [40-50]%;
• Guatemala-Europe: American Airlines: [10-20]%; KLM-Royal Dutch: [5- 10]%; Martinair Holland: [5-10]%; Cargolux: [0-5]%; Livingston: [0-5]%.

209. The market investigation confirmed the existence of alternative credible suppliers in all of the affected routes that would act as a competitive constraint on the parties. A majority of respondents also considered unlikely the possibility that the parties could increase prices following the transaction.

210. One of the competitors that responded to the market investigation expressed a concern about the competitive situation post-merger on intra-European routes and on the routes between Europe and Israel and South Africa. It is to be recalled that the transaction does not lead to affected markets in intra-European routes and that the combined market shares of the parties in the routes Brazil-Europe and Israel-Europe (as the routes Europe-Brazil and Europe-Israel do not give rise to affected markets) are [10- 20]% and [10-20]% respectively. Given the moderate combined market shares of the parties on these routes and the presence of numerous competitors with higher market shares than the parties,92 the transaction does not give rise to competition concerns concerning these routes.

211. One of the customers that responded to the Commission's questionnaire expressed concerns that the transaction may have an adverse effect on direct cargo services  between Europe and Argentina. However, the market investigation confirmed that, in  the air cargo transport sector, any indirect route is a substitute for any direct route and that several alternative credible operators are present on the routes between Europe and Argentina.93

212. In addition to the presence of credible competitors on all of the affected routes, a further competitive constraint on the parties is posed by cargo carriers and freight forwarders that operate the routes indirectly (one-stop). Indeed, the market investigation confirmed that cargo carriers enjoy greater flexibility with regard to schedules and the use of alternative airports and therefore that indirect flights should be included in the relevant cargo market.

213. Finally, the market investigation confirmed that barriers to entry into the air cargo transport markets are low, given the limited costs and time frame required for a  customer to switch to another cargo supplier and the fact that cargo service contracts are normally concluded for one year and do not contain exclusivity clauses. Only some respondents pointed to the existence of traffic rights restrictions which, however, are not specifically related to the air cargo transport sector. In fact, respondents to the market investigation indicated that, in the last three years, several operators entered a number of routes between Europe and some Latin American countries (for example, according to these respondents, LAN entered the routes to Europe from Argentina, Brazil and Chile; Cargo B entered the routes from Europe to Colombia and Venezuela; and Air Europa entered the routes to Europe from Argentina, Chile, Dominican Republic and Peru as well as the routes from Europe to Argentina, Dominican Republic and Venezuela).

214. In view of the foregoing, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to the market for air cargo transport services.

C. Groundhandling services

215. IB provides ground handling services to third parties at airports located in Spain and at the airports of Tel-Aviv (Israel) and Malabo (Equatorial Guinea). BA does  not provide groundhandling services to third parties except for very limited services at the airport of Düsseldorf. Given that BA does not provide groundhandling services in any of the airports where IB offers these services, the transaction leads to no horizontal overlaps on this market.

216. On the other hand, the transaction leads to vertically affected markets as IB provides groundhandling services to BA in five airports in Spain, namely Madrid, Palma de Mallorca, Malaga, Alicante and Ibiza.94

217. The Commission considers, however, that the transaction will not lead to input foreclosure. According to the parties, in the five Spanish airports where IB provides groundhandling services to BA, there are one or more alternative credible suppliers, some of which were market leaders in 2009 in three of the airports concerned:

• Madrid: Swissport International Ltd.: [40-50]%; IB: [30-40]%; Groundforce- Air Europa: [20-30]%;
• Palma de Mallorca: Acciona Airport Services: [60-70]%; IB: [30-40]%;
• Malaga: IB: [40-50]%; Flightcare España: [30-40]%; Clever Handling Services: [20-30]%;
• Alicante: Swissport International Ltd.: [50-60]%; IB: [20-30]%; Flightcare España: [20-30]%;
• Ibiza: IB: [50-60]%; Acciona Airport Services: [40-50]%.

218. The transaction will also not lead to customer foreclosure as BA’s market share as a receiver of groundhandling services in the five airports concerned is below [0-5]%.

219. In view of the foregoing, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to the market for groundhandling services.

D. Maintenance, Repair and Overhaul (MRO) services

220. IB is active on the four segments identified by the Commission within the MRO market, whilst BA is only active on the segments of line maintenance and components maintenance.

221. On the two product segments where the activities of IB and BA overlap, and irrespective of whether the geographic market for each segment is considered to be national, EEA-wide or worldwide, the combined market share of the parties is well below 15%. The proposed transaction does not therefore lead to horizontally affected markets.

222. In addition to the market shares of the parties in each of the MRO segments being limited, the turnover obtained by BA and IB for MRO services provided to each other is very small. For example, the line maintenance services provided by BA to IB in 2009 only represented a [0-5]% of BA's total turnover obtained on this segment, whilst only [0-5]% of IB's total turnover for maintenance services provided to third parties corresponded to BA. The proposed transaction would not therefore lead to any input or customer foreclosure by the parties on any of the segments of the MRO market.

223. In view of the above, the transaction does not raise serious doubts as to its compatibility with the internal market with respect to the market for MRO services.

VII. CONCLUSION

224. For the above reasons, the European Commission has decided not to oppose the notified operation and to declare it compatible with the internal market and with the EEA Agreement. This decision is adopted in application of Article 6(1)(b) of the Merger Regulation.

1  OJ L 24, 29.1.2004, p. 1 ("the Merger Regulation"). With effect from 1 December 2009, the Treaty on 
the Functioning of the European Union ("TFEU") has introduced certain changes, such as the 
replacement of "Community" by "Union" and "common market" by "internal market". The terminology 
of the TFEU will be used throughout this decision. 

2   The oneworld members are: American Airlines, British Airways, Cathay Pacific, Finnair, Iberia, Japan 
Airlines (JAL), LAN, Malév, Mexicana, Qantas, Royal Jordanian. 

3  Turnover calculated in accordance with Article 5(1) of the Merger Regulation.

 4  These three methodologies are explained in Commission Decision of 27 June 2007 in Case No. 
COMP/M.4439 Ryanair/Aer Lingus, OJ C 47, 20.2.2008, p. 9-20, paragraph 13 et seq.

5  COMP/M.5440 Lufthansa/Austrian Airlines; COMP/M.5335 Lufthansa/SN Airholding; COMP/M.5364-
Iberia/Vueling/Clickair; COMP/M.3280 Air France/KLM; COMP/M.3770 Lufthansa/Swiss. 

6  COMP/M.5440 Lufthansa/Austrian Airlines; COMP/M.5335 Lufthansa/SN Airholding; COMP/M.5364-
Iberia/Vueling/Clickair; COMP/M.3280 Air France/KLM; COMP/M.3770 Lufthansa/Swiss.  In the 
present case, time sensitive and non-time sensitive passengers have been distinguished on the basis of 
CAA and MIDT data (see Section VI.A.1 and A.2). 

7  Replies to the questionnaires to corporate customers (questions 9 et seq.), travel agents (questions 10 et seq.) and competitors (questions 12 et seq.). 

8  Replies to the questionnaires to competitors, corporate customers and travel agents. 

9  Average revenue per ticket is used as a proxy for the actual fare paid. 

10  "Non-stop" flights are flights that take off at airport A and land at airport B, where they offload passengers, and that are constantly in the air from their origin airport and to their final destination airport. By contrast, "direct" flights may entail a refuelling stop and/or a disembarking/re-embarking stop, but are marketed under a single flight code and are flown with a single aircraft. "One-stop" flights include direct flights that do not qualify as "non-stop", as well as indirect flights which are journeys that require a change of aircraft or a change of flight code. Unless indicated otherwise, the terms "non-stop/one-stop" and "direct/indirect" are used interchangeably in the present decision as this terminology does not have an effect on the outcome of the competitive assessment in the present case. 

11  COMP/M.5440 Lufthansa/Austrian Airlines; COMP/M.2041 United/US Airways; COMP/M.2672 
SAS/Spanair. 

12  COMP/M.5440 Lufthansa/Austrian Airlines, COMP/M.5335 Lufthansa/SN Airholding; COMP/M. 5384 
Iberia/Vueling/Clickair. Various elements such as price, level of frequencies and the alternative services 
offered play a role in the level of substitutability of indirect flights to direct services. 

13  COMP/M.5440 Lufthansa/Austrian Airlines, COMP/M.2041 United/US Airways, COMP/M.2672 
SAS/Spanair. 

14  COMP/M.5440 Lufthansa/Austrian Airlines; COMP/M.5335 Lufthansa/SN Airholding; COMP/M.3280 
Air France/KLM. . See also judgment of the General Court of 6 July 2010 in case T-342/07 Ryanair v 
Commission paras. 99 et seq. 

15  COMP/M. 967 KLM/Air UK, para. 24; COMP/D2/38479 British Airways/Iberia/GB Airways, para. 21-24 
of the summary of the Commission's assessment published on 10 January 2004. In COMP/M.4439 
Ryanair/Aer Lingus, London Heathrow, Gatwick, Stansted, Luton and City airports were considered 
substitutable for scheduled point-to-point passenger air transport services between Dublin and London. 

16  See e.g. replies to question 24 of the questionnaire to travel agents. 

17  A similar pattern is observed for business fully flexible tickets. 

18  Replies to questions 27 et seq. of the questionnaire to corporate customers, questions 21 et seq. of the 
questionnaire to travel agents and questions 22 et seq. of the questionnaire to competitors. 

19  UK Competition Commission BAA Airports market investigation, Final Report, 19 March 2009. 

20  See replies to questions 22 et seq. of the questionnaire to airports and to questions 6 et seq. to CAA. 

21  COMP/D2/38479 British Airways/Iberia/GB Airways, para. 21-24 of the summary of the Commission's 
assessment published on 10 January 2004 (see in more detail Section V.B below). In the absence of 
relevant air services between LCY and Spain, a substitutability of LHR and LGW with LCY was not 
discussed in that decision. 

22  See e.g. questions 26 et seq. of the questionnaire to travel agents. This result is in line with 
COMP/M.5364 Iberia/Vueling/Clickair where the Commission found that El Prat, Girona and Reus were 
not substitutable for a number of Spanish domestic and European routes, at least for time sensitive 
passengers. 

23  COMP/M.4439 Ryanair/Aer Lingus, COMP/M.5141 KLM/Martinair. 

24  See in more detail section VI.A.2.1 below. 

25  COMP/M.1328 KLM/Martinair; Case IV/35.545 Lufthansa/SAS; COMP/M.3280 Air France/KLM; 
COMP/M.5440 Lufthansa/Austrian Airlines.

 26  Integrators offer not only airport to airport services but also handle cargo from the point of origin to the 
airport and from the airport to the final destination, including legal formalities such as customs clearance. 
Examples are DHL, UPS, FedEx and TNT. 

27  COMP/M.5141 KLM/Martinair; COMP/M.5181 Delta Air Lines/Northwest Airlines; COMP/M.5335 
Lufthansa/SN, COMP/M.5440 Lufthansa/Austrian Airlines.  

28  COMP/M.3280 Air France/KLM, COMP/M.5141 KLM/Martinair, COMP/M.5440 Lufthansa/Austrian 
Airlines.

29  COMP/JV.19 KLM/Alitalia; Case No IV/M.1506 Singapore Airlines/Rolls-Royce; COMP/M.3280 Air 
France/KLM; COMP/M.5440 Lufthansa/Austrian Airlines.

 30  COMP/M.3280 Air France/KLM; COMP/M.3770 Lufthansa/Swiss; COMP/M.5440 Lufthansa/Austrian 
Airlines.  

31  Council Directive 96/67/EC of 15 October 1996 on access to the groundhandling market at Community 
airports, OJ L 272, 25.10.1996, pp. 36-45. 

32  COMP/M.2254 Aviapartner/Maersk/Novia; Case IV/M.1913 Lufthansa/Menzies/LGS/JV; COMP/M.5364 
Iberia/Vueling/Clickair; COMP/M.5440 Lufthansa/Austrian Airlines.

33  Case IV/M. 786 Birmingham International Airport; Case IV/M. 1035 Hochtief/Aer Rianta/Düsseldorf 
Airport; COMP/M.2262 Flughafen Berlin II; COMP/M.3382 Iberia/Cobra Instalaciones/Serpista JV;
COMP/M.3823 MAG/Ferrovial Aeropuertos/Exeter Airport; COMP/M.4164 Ferrovial/ 
Quebec/GIC/BAA.

34  COMP/M.1913 Lufthansa/Menzies/LGS/JV; COMP/M.2262 Flughafen Berlin II; COMP/M.4164 
Ferrovial/Quebec/GIC/BAA. 

35  Case IV/M.1255 Privatisation, Berlin airport; COMP/M.2315 The Airline Group/NATS. 

36  COMP/JV.19 KLM/Alitalia, COMP/M.3280 Air France/KLM, COMP/M.3374 SR Technics/FLS 
Aerospace, COMP/M.5440 Lufthansa/Austrian Airlines. 

37  COMP/M.3374 SR Technics/FLS Aerospace. 

38  COMP/M.3280 Air France/KLM. 

39  COMP/JV.19 KLM/Alitalia. 

40 COMP/M.5399 Mubadala/Rolls Royce/JV; COMP/M.5403 Lufthansa/bmi; COMP/M.5440 
Lufthansa/Austrian Airlines.

41  Cooperation includes areas such as check-in, lounges, boarding experience, cabin crew, meals, seat 
comfort, punctuality, in-flight entertainment, aircraft cleanliness and baggage handling. They have 
combined ticket offices, check-in facilities and globally provide almost 550 lounges. Some members, 
including the parties, also code-share with each other on certain routes and cooperate in relation to 
Frequent Flyer Programmes ("FFP"). 

42  COMP/M.5440 Lufthansa/Austrian Airlines, COMP/M.5335 Lufthansa/SN.

43  Additionally, BA acquired a 9% indirect stake in Iberia. Currently, BA has a stake of 13.15% in Iberia, 
while Iberia has a stake of 9.07% in BA. 

44  In addition, on 20 December 2002, BA and Iberia signed a Trilateral Alliance Agreement with GB 
Airways. Both agreements were notified to the Commission by applying for an exemption under Article 5 
of former Regulation Nº 3975/87 (see case COMP/D2/38.479 BA/IB/GB). This agreement came to an end 
when GB was taken over by easyJet in 2008. 

45  See section 7 of the exemption decision (case COMP/D2/38.479 BA/IB/GB), i.e. on the routes London-
Madrid, London-Barcelona, London-Bilbao, London-Seville and London-Valencia. In particular, the 
parties offered to divest some slots at both ends on LGW-Madrid (which were eventually taken by 
easyJet), while on London-Barcelona a frequency freeze for a minimum period of two IATA seasons was 
considered sufficient to remedy all competition concerns.

 46  Under the new procedural rules for the application of Article 101 TFEU, notably Regulation 1/2003, it 
was not possible for the parties to apply for a new exemption. [Other confidential information][…]

47  In addition, easyJet operates three daily frequencies from LTN to Madrid. 

48  The Commission cannot be required to accept as a counterfactual a pre-merger cooperation that is 
contrary to Article 101 TFEU, see cases COMP/M.5403 Lufthansa/bmi para. 42, COMP/M.5335 
Lufthansa/SN para. 263 and COMP/M.5440 Lufthansa/Austrian Airlines para. 106. Indeed, if the possible 
illegality of a pre-merger agreement between the parties could not be taken into account in the context of 
merger control, the parties could argue that there would only be a small reduction or even no reduction of 
competition as a result of the merger. A merger decision in such circumstances would effectively 
incorporate and perpetuate the pre-merger illegality for ever, since mergers that are approved under the 
Merger Regulation are no longer challengeable under Article 101 TFEU.

 
49  On LON-BCN, the present level of frequencies operated by easyJet matches the level of frequencies 
operated by easyJet and that was deemed sufficient at the time of the exemption decision.

50  No competition concerns are deemed to arise on all affected direct-indirect overlaps where throughout the last four IATA seasons (i) the parties' combined market share was below 25%, or (ii) one of the parties 
had a market share below 2%, or (iii) as regards short-haul routes where the total share of indirect 
operations in the relevant market was below 10%, or (iv) as regards transatlantic routes on which IB, BA 
and AA are active and have concluded a joint-venture agreement where the combined market share of 
IB/BA/AA is below 25% or the market share of either IB or BA is below 2%. Similarly, no competition 
concerns are deemed to arise on all indirect-indirect overlaps where throughout the last four IATA 
seasons (i) the parties’ combined market share was below 25% or (ii) one of the parties had a market 
share below 2%, or (iii) as regards short-haul routes where the total annual traffic was below 15,000 
passengers, or (iv) as regards long-haul routes where the total annual traffic was below 30,000 
passengers. 

51  As mentioned in section IV.A.3, the terms "direct" and "non-stop" are used interchangeably in the present 
case, unless indicated otherwise. 

52  As of the winter season 09/10 BA stopped operating from LGW on this route. Furthermore, within the 
next few months, the parties expect to implement a code-share agreement pursuant to which IB will put its 
code on Madrid – LCY flights operated by BA City Flyer. 

53  EasyJet is not maintaining a level of 4 daily frequencies during all Summer 2010 (e.g. for the period 15-
21 July 2010, it offers only 3 daily frequencies). EasyJet operates further 3 daily frequencies from LTN. 

54  Ryanair operates further 2 daily frequencies from STN.  

55   On a market encompassing flights between all five London airports and Madrid airport, the parties had a 
combined market share of [60-70]% in Summer 09 and [50-60]% in Winter 09/10 on the time sensitive 
passenger market, while easyJet had a market share of [20-30]% and [30-40]%, Air Europa had a market 
share of [0-5]% and [0-5]%, and Ryanair had a market share of [5-10]% and [5-10]% respectively (total 
number of time sensitive passengers in Winter 09/10: [150,000-200,000]). As regards non-time sensitive 
passengers, the parties had a combined market share of [30-40]% in Summer 09 and [20-30]% in Winter 
09/10, while easyJet had a market share of [30-40]% in both seasons, Air Europa had a market share of 
[10-20]% in both seasons and Ryanair had a market share of [20-30]% and [20-30]% respectively (total 
number of non-time sensitive passengers in Winter 09/10: [450,000 – 500,000]). 

56  The market share are based inter alia on CAA data, tracking the number of passengers flown on the each 
respective route. CAA also conducts a survey identifying several groups of passengers based on the 
purpose of their journey (business or leisure) which are in the table above used as proxies for time 
sensitive and non-time sensitive passengers respectively. 

57  The term "base" may be used to characterise airports on which airlines base their aircraft and on which 
they concentrate their operations, offering mainly flights from and to these "base" airports, see 
COMP/M.4439 Ryanair/Aer Lingus para. 44 

58  Replies to question 36 of the questionnaire to corporate customers. This is in line with the findings in the 
2003 exemption decision, see para. 28. 

59  Reply to question 12 of the questionnaire to corporate customers. 

60  Submission of the parties of June 2010. 

61   The fact that easyJet is a credible option for time sensitive passengers is further in line with previous 
Commission decision practice, see COMP/M.5335 Lufthansa/SN paras. 217 et seq. Furthermore, the UK 
CAA study of May 2009 on UK Business Air Travel which relates to the use of air services by business 
travellers found that Ryanair also transports business passengers when flying to "primary" airports (like 
Madrid)

62  Replies of question 38 of the questionnaire to corporate customers and question 29 of the questionnaire to travel agents. 

63  E.g. replies to question 45 of the questionnaire to corporate customers. 

64  See http://business.easyjet.com. 

65  Replies of question 37 of the questionnaire to corporate customers and question 28 of the questionnaire to travel agents. 

66  Replies to question 56 of the questionnaire to corporate customers. See also case COMP/M.5335 
Lufthansa/SN Airholding, para. 202 et seq. 

67  Ryanair offers 1 daily flight in the evening.

68  Replies to question 31 of the questionnaire to corporate customers and to question 24 of the questionnaire to travel agents, reply to questionnaire to competitors. 

69  This is in line with indications revealed in the market investigation that all London airports are 
substitutable for non-time sensitive passengers, see section IV.A.4.1 above.

 70  Some competitors have complained about the negative effects of the transaction at LHR and that the 
transaction leads to a further concentration of slots at that airport. These concerns are mitigated by the 
results of the market investigation according to which LHR, LCY and LGW are sufficiently substitutable. 
Moreover, the increment brought about by the merger, ie. [0-5]% of LHR slot held by IB, is not 
significant in the context of our analyses in the present case.

 71  At LGW, LCY and Madrid, constraints appear to exist in the key morning and evening times; replies to 
questionnaires to airports and slot coordinators.

72  Reply to question 34 of the questionnaire to competitors. 

73  One further slot at LGW obtained by easyJet in the framework of the antitrust commitments still remains 
with easyJet but easyJet no longer uses that slot for services on the Madrid-London route. Moreover, 
easyJet holds the largest slot portfolio at LGW.

74  For the reasons detailed above, it is also unlikely that any of the concerns expressed by respondents in the course of the market investigation (relating in particular to possible price increases, capacity reductions or 
anti-competitive practices vis-à-vis travel agents) will materialise post-merger. More in particular with 
regard to travel agents, despite some concerns voiced as regards the strong position of the merged entity 
vis-à-vis travel agents, it should be noted that many travel agents do no expect a significant negative 
effect of the transaction on the level of commission/other conditions offered to them by the parties, see 
replies to question 3 of the questionnaire to travel agents. Furthermore, services of both easyJet and Air 
Europa are available via GDS and thus also available to travel agents. 

75   The code-share agreements between the parties on London-Barcelona (as well as on London-Madrid) is 
[description of the code-share agreement] […] 

76  Frequency data refer to Winter 09/10. It should be noted that easyJet is offering 4 daily frequencies during Summer 2010 (reference period 15-21 July 2010). 

77   On a market encompassing flights between all five London airports and El Prat, the parties had a 
combined market share of [50-60]% in Summer 09 and [50-60]% in Winter 09/10 on the time sensitive 
market, while easyJet had a market share of [40-50]% and [40-50]% (total number of time sensitive 
passengers in Winter 09/10: [150,000 – 200,000]). As regards non-time sensitive passengers, the parties 
had a combined market share of [40-50]% in Summer 09 and [30-40]% in Winter 09/10, while easyJet 
had a market share of [50-60]% and [60-70]% (total number of non-time sensitive passengers in Winter 
09/10: [200,000-300,000]). 

78   http://www.air-journal.fr/2010-06-01-easyjet-veut-aussi-sa-base-a-barcelone-57369.html. 

79  See p. 41 and 42 of the report.

80  Replies to question 31 of the questionnaire to corporate customers and to question 24 of the questionnaire to travel agents, reply to questionnaire to competitors. the study of the UK CAA of May 2009 found that on the route London-Barcelona, easyJet's LTN service carried a similar proportion of business passengers to IB's LHR flights, see p. 41 and 42 of the report. 

81  See also replies to question 28 of the questionnaire to travel agents. 

82  [IB's strategic considerations concerning the route] 

83  For the reasons detailed above, as for the London-Madrid route, it is also unlikely that any of the concerns expressed by respondents in the course of the market investigation (relating in particular to possible price 
increases, capacity reductions or anti-competitive practices vis-à-vis travel agents) will materialise post-
merger. 

84  Without prejudice of airport substitutability at the London end of the relevant route as the non-stop 
services provided by easyJet and Ryanair out of London airports different from those where the parties 
operate provide a sufficient competitive constraint to IB/BA operations. This is evidenced by the large 
number of passengers on easyJet's and Ryanair's flights on these routes and it has been confirmed by the 
market investigation for the thickest route (London-Malaga).

85  EasyJet and Ryanair are both present on the four UK-Spain routes concerned. On the thickest routes 
(London-Alicante and London-Malaga) also Monarch is present. 

86  The market share of Air France /KLM is [30-40]% in Summer 09 and [40-50]% in Winter 09/10, 
disregarding non-MIDT sales. If account is taken of these sales its market share is [20-30]% in Summer 
09 and [5-10]% in Winter 09/10. 

87  No relevant overlap between the parties exists as far as time sensitive passengers are concerned.

88  As indicated by the parties, IB places its code on Mexicana operated flights from Mexico City to several 
domestic destinations in connection with its own operated flights between Madrid and Mexico City. IB 
also places its code on the routes Mexico-Guatemala, operated by Mexicana, and Miami-Cancun in 
connection with IB’s Madrid-Miami flight. Mexicana places its code on flights operated by IB from Madrid
to domestic and European destinations in connection with its own-operated flights between Mexico and 
Madrid. However, currently there is no code share agreement on the Madrid-Mexico City route. A previously 
existing code share agreement between IB and Mexicana was cancelled on [date][…], i.e. shortly after 
Mexicana started its direct operation on this route. Therefore, Mexicana is to be considered as an active 
competitor of the merged entity on this route, even if account is taken of the fact that it also belongs to the 
oneworld alliance.

89 The joint venture between United and Aer Lingus was announced in January 2009, the operation did not 
commence until March 2010.  Only United Airlines sales are reflected in the table (both non-stop and 
one-stop sales) because Aer Lingus' market shares are not reflected in the MIDT data during the Winter 
season 09/10. 

90 TAM is a Brazilian carrier and member of Star Alliance.

91 Includes Mexico and the Caribbean (excludes French Guyana).

92  On the route Brazil-Europe: Lufthansa: [20-30]%; TAP Air Portugal: [20-30]%; Air France: [10-20]%; 
Cargolux: [5-10]%; KLM Royal Dutch: [5-10]%; Swiss International Airlines: [5-10]%; Alitalia: [0-5]%. 
On the route Israel-Europe: Lufthansa: [30-40]%; Swiss International Airlines: [20-30]%; British 
Midland: [10-20]%; Austrian Airways: [0-5]%; Air France: [0-5]%; Alitalia: [0-5]%; others: [0-5]%.

93  On the route Argentina-Europe: Air France: [20-30]%; Lufthansa: [10-20]%; Alitalia: [10-20]%; 
Aerolineas Argentinas: [10-20]%. On the route Europe-Argentina: Air France: [20-30]%; Lufthansa: [20-
30]%; Aerolineas Argentinas: [10-20]%; LAN Chile: [5-10]%; Alitalia: [5-10]%. 

94  IB has also provided during the last three years negligible groundhandling services to BA in four other 
airports, namely Girona ([10-15] aircraft assisted), Tenerife Sur ([5-10] aircraft), Valencia ([0-5] aircraft) 
and Vitoria ([0-5] aircraft).